This article has its origins in a generalized feeling of dissatisfaction with current theories about political development trends in Latin America. It is an early statement of a series of arguments which will subsequently be developed in a forthcoming book.

The veritable explosion of empirically grounded monographs in the last fifteen or twenty years has made the task of producing a synthetic account of Latin American development simultaneously more pressing and more difficult: more difficult because it has made simple explanatory models harder to sustain, and has opened up the accepted historiography to serious and widespread revisionist attack; more pressing because many, if not most, social scientists accept the need to develop a theory of social change which is historically grounded, capable of explaining large-scale social transformations. My concern in this article is with the methodological issues involved in the formulation of an adequate theory of Latin American development, rather than with establishing new facts. There is considerable historiographical controversy over many of the events discussed in this article, and in these cases I have made my own judgement about where, on balance, the evidence points.

Exporters of raw materials under Iberian rule, the nations of Latin America continued to perform a similar role in the world economy after Independence. In the nineteenth century, however, a significant shift occurred in the kind of materials exported. Whereas in colonial times the great wealth of Latin America lay in her mineral resources, particularly silver and gold, aster 1850 agricultural production for foreign markets took on larger importance. The export of foodstuffs was not a new phenomenon, but in the nineteenth century the growth in consumer demand in the industrializing nations and the developing revolution in. transport much enhanced the incentives for Latin Americans who would produce coffee, wheat, cattle, or bananas for overseas markets.

Like any major historical phenomenon, the Mexican Revolution can be viewed from a variety of angles. From one, arguably the most important, it was a rural phenomenon, rightly categorised by Eric Wolf as a ‘peasant war’, hence comparable to the Russian or Chinese Revolutions. Form another it can be seen as a generalised social and political (some might like to call it ‘hegemonic’ crisis, marking the end of the old oligarchic Porfirian order and characterised by mass political mobilisation; as such it bears comparison with the crises experienced in Italy and Germany after the First World War; in Spain in the early 1930s; in Brazil in the 1960s or Chile in the 1970s. But what it emphatically was not was a workers' revolution. No Soviets or workers' party sought — let alone attained — political hegemony. No Soviets or workers' councils were established, as in Petrograd or Berlin. There were no attempts at works' control of industry, as in Turin, Barcelina — or the gran mineria of Bolivia.

The origins of the labour movement in Argentina, which date from a period when the country was developing rapidly and had an important place in the world economy, have received little close attention from historians. Generally, the years before 1930 have been neglected in favour of the study of developments associated with Perón and Peronism. Such secondary literature as does exist consists largely of personal memoirs, whose authors were, understandably, parti pris. The more general studies tend to underestimate the extent, variety and sophistication of trade unionism before 1930, and also to contrast the style and ideology of the unions with that of the Peronist labour movement. Most notably, Argentina is widely quoted, along with Spain, as one of the countries where anarchist and/or anarcho-syndicalist ideology and practice dominated the labour movement.

The country is so well-favoured that if it were rightly cultivated it would yield everything. (Letter of Pero Vaz de Caminha, 1 May 1500.)

It is a vast region with favoured terrain. On its soil grow all fruits; On its subsoil exist all treasures… Its fields give the most useful food; its mines the finest gold… It is an admirable country, rich in every respect, where prodigiously profuse nature sacrifices herself in fertile produce for the opulence of the monarchy and the benefit of the world. (J. da Rocha Pitta, 1724.)

The New International Economic Order is an ill-defined concept which is nonetheless of considerable political importance. Non-existing in present-day international reality, it represents a vision, especially on the part of developing countries, of a future in which the latter will be an integral part of the world economy, instead of existing at its margins. As an idea, the New International Economic Order was given formal expression in 1974, though its historical roots go back to at least 1960.

In a recent article in this journal, Dr Bergad argues that the production of coffee for the export economy in nineteenth-century Puerto Rico was accompanied by the emergence of a rural proletariat, and suggests that the latter process in fact predated the development of a sugar plantation economy in the twentieth century, with which capitalist free wage labour is commonly associated. This characterization of the coffee hacienda workforce raises definitional problems on which I would like to comment. This characterization of the coffee hacienda workforce raises definitional problems on which I would like to comment. Illustrating with examples from coffee-producing areas in Latin America and sugar-producing areas in the Caribbean, I shall argue that the increased demand for labour-power arising from the expansion of an export-oriented yet labour-intensive capitalist agriculture in a location where labour is scarce is met not by free wage labour but rather by recourse to unfree labour. In these circumstances, competition for labour-power results paradoxically in the attempt by capitalists to restrict the free movement of labour and the consequent formation of a free labour market.

In his comment on my article ‘Coffee and Rural Proletarianization in Puerto Rico, 1840–1898’ (Journal of Latin American Studies, XV, no. 1, May 1983, pp. 83–100), Tom Brass is to be lauded for the comparative observations used to contest my conclusions on proletarianization in the coffee-producing regions of 19th-century Puerto Rico. By examining some of the literature on coffee expansion in Brazill and Colombia, and comparing these cases with Puerto Rico, Dr Brass concludes that the development of capitalist agriculture, when accompanied by labour scarcity, results in ‘unfree labour’ rather than the development of free wage labour as I have indicated to be the case for Puerto Rico' coffee sector.